/* Test for NUL byte processing via empty string. */
static int
utf8_test_2 (void)
{
wchar_t wc;
mbstate_t s;
wc = 42; /* arbitrary number */
memset (&s, 0, sizeof (s)); /* get s into initial state */
assert (mbrtowc (NULL, "", 1, &s) == 0); /* valid terminator */
assert (mbsinit (&s));
wc = 42; /* arbitrary number */
memset (&s, 0, sizeof (s)); /* get s into initial state */
assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
assert (mbrtowc (NULL, "", 1, &s) == (size_t) -1); /* invalid terminator */
wc = 42; /* arbitrary number */
memset (&s, 0, sizeof (s)); /* get s into initial state */
assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
assert (mbrtowc (&wc, "\x89", 1, &s) == (size_t) -2); /* 2nd byte processed */
assert (mbrtowc (NULL, "", 1, &s) == (size_t) -1); /* invalid terminator */
wc = 42; /* arbitrary number */
memset (&s, 0, sizeof (s)); /* get s into initial state */
assert (mbrtowc (&wc, "\xE2", 1, &s) == (size_t) -2); /* 1st byte processed */
assert (mbrtowc (&wc, "\x89", 1, &s) == (size_t) -2); /* 2nd byte processed */
assert (mbrtowc (&wc, "\xA0", 1, &s) == 1); /* 3nd byte processed */
assert (mbrtowc (NULL, "", 1, &s) == 0); /* valid terminator */
assert (mbsinit (&s));
return 0;
}
void test_mbrtowc_incomplete(mbstate_t* ps) {
ASSERT_STREQ("C.UTF-8", setlocale(LC_CTYPE, "C.UTF-8"));
uselocale(LC_GLOBAL_LOCALE);
wchar_t out;
// 2-byte UTF-8.
ASSERT_EQ(static_cast<size_t>(-2), mbrtowc(&out, "\xc2", 1, ps));
ASSERT_EQ(1U, mbrtowc(&out, "\xa2" "cdef", 5, ps));
ASSERT_EQ(static_cast<wchar_t>(0x00a2), out);
ASSERT_TRUE(mbsinit(ps));
// 3-byte UTF-8.
ASSERT_EQ(static_cast<size_t>(-2), mbrtowc(&out, "\xe2", 1, ps));
ASSERT_EQ(static_cast<size_t>(-2), mbrtowc(&out, "\x82", 1, ps));
ASSERT_EQ(1U, mbrtowc(&out, "\xac" "def", 4, ps));
ASSERT_EQ(static_cast<wchar_t>(0x20ac), out);
ASSERT_TRUE(mbsinit(ps));
// 4-byte UTF-8.
ASSERT_EQ(static_cast<size_t>(-2), mbrtowc(&out, "\xf0", 1, ps));
ASSERT_EQ(static_cast<size_t>(-2), mbrtowc(&out, "\xa4\xad", 2, ps));
ASSERT_EQ(1U, mbrtowc(&out, "\xa2" "ef", 3, ps));
ASSERT_EQ(static_cast<wchar_t>(0x24b62), out);
ASSERT_TRUE(mbsinit(ps));
// Invalid 2-byte
ASSERT_EQ(static_cast<size_t>(-2), mbrtowc(&out, "\xc2", 1, ps));
ASSERT_EQ(static_cast<size_t>(-1), mbrtowc(&out, "\x20" "cdef", 5, ps));
ASSERT_EQ(EILSEQ, errno);
}
void
ui_init(bool force_ascii)
{
// Cell character 0 is always a space
strcpy(ui_chars[0], " ");
#ifdef __STDC_ISO_10646__
if (force_ascii) {
ui_ascii = true;
return;
}
// Encode Unicode cell characters using system locale
char *origLocale = setlocale(LC_CTYPE, NULL);
setlocale(LC_CTYPE, "");
int ch;
mbstate_t mbs;
memset(&mbs, 0, sizeof mbs);
for (ch = 1; ch < NCHARS; ch++) {
int len = wcrtomb(ui_chars[ch], 0x2580 + ch, &mbs);
if (len == -1 || !mbsinit(&mbs)) {
ui_ascii = true;
break;
}
ui_chars[ch][len] = 0;
}
// Restore the original locale
setlocale(LC_CTYPE, origLocale);
#else
ui_ascii = true;
#endif
}
size_t
wcrtomb (char *s, wchar_t wc, mbstate_t *ps)
{
/* This implementation of wcrtomb on top of wctomb() supports only
stateless encodings. ps must be in the initial state. */
if (ps != NULL && !mbsinit (ps))
{
errno = EINVAL;
return (size_t)(-1);
}
if (s == NULL)
/* We know the NUL wide character corresponds to the NUL character. */
return 1;
else
{
int ret = wctomb (s, wc);
if (ret >= 0)
return ret;
else
{
errno = EILSEQ;
return (size_t)(-1);
}
}
}
int test_newbuf(void)
{
Buf *buf = newbuf();
/* newbuf sets shiftstate to initial shift state */
assert(mbsinit(&buf->shiftstate));
return 0;
}
int main( void )
{
mbstate_t mbs;
memset(&mbs, 0, sizeof mbs);
TESTCASE(mbsinit(NULL) != 0);
TESTCASE(mbsinit(&mbs) != 0);
#ifndef REGTEST
// Surrogate pending
mbs._Surrogate = 0xFEED;
TESTCASE(mbsinit(&mbs) == 0);
mbs._Surrogate = 0;
mbs._PendState = 1;
TESTCASE(mbsinit(&mbs) == 0);
#endif
return TEST_RESULTS;
}
TInt CTestLibcwchar::testmbsinit2L()
{
int retVal = mbsinit(NULL);
if(retVal != 0)
{
return KErrNone;
}
return KErrGeneral;
}
TEST(wchar, wcrtomb_start_state) {
char out[MB_LEN_MAX];
mbstate_t ps;
// Any non-initial state is invalid when calling wcrtomb.
memset(&ps, 0, sizeof(ps));
EXPECT_EQ(static_cast<size_t>(-2), mbrtowc(NULL, "\xc2", 1, &ps));
EXPECT_EQ(static_cast<size_t>(-1), wcrtomb(out, 0x00a2, &ps));
EXPECT_EQ(EILSEQ, errno);
// If the first argument to wcrtomb is NULL or the second is L'\0' the shift
// state should be reset.
memset(&ps, 0, sizeof(ps));
EXPECT_EQ(static_cast<size_t>(-2), mbrtowc(NULL, "\xc2", 1, &ps));
EXPECT_EQ(1U, wcrtomb(NULL, 0x00a2, &ps));
EXPECT_TRUE(mbsinit(&ps));
memset(&ps, 0, sizeof(ps));
EXPECT_EQ(static_cast<size_t>(-2), mbrtowc(NULL, "\xf0\xa4", 1, &ps));
EXPECT_EQ(1U, wcrtomb(out, L'\0', &ps));
EXPECT_TRUE(mbsinit(&ps));
}
// -----------------------------------------------------------------------------
//Function Name :testmbinit3
//API Tested :mbsinit
//TestCase Description:mbrinit returns -> zero if mbstate_t state is not a null
// pointer.
// -----------------------------------------------------------------------------
TInt CTestLibcwchar::testmbsinit3L()
{
mbstate_t state;
int retVal = mbsinit(&state);
if(retVal == 0)
{
return KErrNone;
}
return KErrGeneral;
}
TInt CTestLibcwchar::testmbsinit1L()
{
mbstate_t state;
memset(&state,0,sizeof(mbstate_t));
int retVal = mbsinit(&state);
if(retVal != 0)
{
return KErrNone;
}
return KErrGeneral;
}
extern size_t
tuklib_mbstr_width(const char *str, size_t *bytes)
{
const size_t len = strlen(str);
if (bytes != NULL)
*bytes = len;
#if !(defined(HAVE_MBRTOWC) && defined(HAVE_WCWIDTH))
// In single-byte mode, the width of the string is the same
// as its length.
return len;
#else
mbstate_t state;
memset(&state, 0, sizeof(state));
size_t width = 0;
size_t i = 0;
// Convert one multibyte character at a time to wchar_t
// and get its width using wcwidth().
while (i < len) {
wchar_t wc;
const size_t ret = mbrtowc(&wc, str + i, len - i, &state);
if (ret < 1 || ret > len)
return (size_t)-1;
i += ret;
const int wc_width = wcwidth(wc);
if (wc_width < 0)
return (size_t)-1;
width += wc_width;
}
// Require that the string ends in the initial shift state.
// This way the caller can be combine the string with other
// strings without needing to worry about the shift states.
if (!mbsinit(&state))
return (size_t)-1;
return width;
#endif
}
void TextTerminal::PutChar(TextBuffer* textbuf, char c)
{
if ( ansimode )
return PutAnsiEscaped(textbuf, c);
if ( mbsinit(&ps) )
{
switch ( c )
{
case '\n': Newline(textbuf); return;
case '\r': column = 0; return;
case '\b': Backspace(textbuf); return;
case '\t': Tab(textbuf); return;
case '\e': AnsiReset(); return;
case 127: return;
default: break;
}
}
wchar_t wc;
size_t result = mbrtowc(&wc, &c, 1, &ps);
if ( result == (size_t) -2 )
return;
if ( result == (size_t) -1 )
{
memset(&ps, 0, sizeof(ps));
wc = L'�';
}
if ( result == (size_t) 0 )
wc = L' ';
if ( textbuf->Width() <= column )
Newline(textbuf);
TextPos pos(column++, line);
TextChar tc(wc, vgacolor, ATTR_CHAR | next_attr);
textbuf->SetChar(pos, tc);
next_attr = 0;
}
/* Returns the number of columns needed to represent the multibyte
character string pointed to by STRING of length NBYTES. If a
non-printable character occurs, and MBSW_REJECT_UNPRINTABLE is
specified, -1 is returned. */
int
mbsnwidth (const char *string, size_t nbytes, int flags)
{
const char *p = string;
const char *plimit = p + nbytes;
int width;
width = 0;
if (MB_CUR_MAX > 1)
{
while (p < plimit)
switch (*p)
{
case ' ': case '!': case '"': case '#': case '%':
case '&': case '\'': case '(': case ')': case '*':
case '+': case ',': case '-': case '.': case '/':
case '0': case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
case ':': case ';': case '<': case '=': case '>':
case '?':
case 'A': case 'B': case 'C': case 'D': case 'E':
case 'F': case 'G': case 'H': case 'I': case 'J':
case 'K': case 'L': case 'M': case 'N': case 'O':
case 'P': case 'Q': case 'R': case 'S': case 'T':
case 'U': case 'V': case 'W': case 'X': case 'Y':
case 'Z':
case '[': case '\\': case ']': case '^': case '_':
case 'a': case 'b': case 'c': case 'd': case 'e':
case 'f': case 'g': case 'h': case 'i': case 'j':
case 'k': case 'l': case 'm': case 'n': case 'o':
case 'p': case 'q': case 'r': case 's': case 't':
case 'u': case 'v': case 'w': case 'x': case 'y':
case 'z': case '{': case '|': case '}': case '~':
/* These characters are printable ASCII characters. */
p++;
width++;
break;
default:
/* If we have a multibyte sequence, scan it up to its end. */
{
mbstate_t mbstate;
memset (&mbstate, 0, sizeof mbstate);
do
{
wchar_t wc;
size_t bytes;
int w;
bytes = mbrtowc (&wc, p, plimit - p, &mbstate);
if (bytes == (size_t) -1)
/* An invalid multibyte sequence was encountered. */
{
if (!(flags & MBSW_REJECT_INVALID))
{
p++;
width++;
break;
}
else
return -1;
}
if (bytes == (size_t) -2)
/* An incomplete multibyte character at the end. */
{
if (!(flags & MBSW_REJECT_INVALID))
{
p = plimit;
width++;
break;
}
else
return -1;
}
if (bytes == 0)
/* A null wide character was encountered. */
bytes = 1;
w = wcwidth (wc);
if (w >= 0)
/* A printable multibyte character. */
{
if (w > INT_MAX - width)
goto overflow;
width += w;
}
else
/* An unprintable multibyte character. */
if (!(flags & MBSW_REJECT_UNPRINTABLE))
{
if (!iswcntrl (wc))
{
if (width == INT_MAX)
goto overflow;
width++;
}
}
else
return -1;
p += bytes;
}
while (! mbsinit (&mbstate));
}
break;
}
return width;
}
while (p < plimit)
{
unsigned char c = (unsigned char) *p++;
if (isprint (c))
{
if (width == INT_MAX)
goto overflow;
width++;
}
else if (!(flags & MBSW_REJECT_UNPRINTABLE))
{
if (!iscntrl (c))
{
if (width == INT_MAX)
goto overflow;
width++;
}
}
else
return -1;
}
return width;
overflow:
return INT_MAX;
}
int
main(int argc, char *argv[])
{
mbstate_t state;
wchar_t wc;
size_t ret;
int mode;
/* configure should already have checked that the locale is supported. */
if (setlocale(LC_ALL, "") == NULL) {
fprintf(stderr, "unable to set standard locale\n");
return 1;
}
/* Test NUL byte input. */
{
const char *src;
memset(&state, '\0', sizeof(mbstate_t));
src = "";
ret = mbsnrtowcs(NULL, &src, 1, 0, &state);
assert(ret == 0);
assert(mbsinit (&state));
src = "";
ret = mbsnrtowcs(NULL, &src, 1, 1, &state);
assert(ret == 0);
assert(mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
src = "";
ret = mbsnrtowcs(&wc, &src, 1, 0, &state);
assert(ret == 0);
assert(wc == (wchar_t) 0xBADFACE);
assert(mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
src = "";
ret = mbsnrtowcs(&wc, &src, 1, 1, &state);
assert(ret == 0);
assert(wc == 0);
assert(mbsinit (&state));
}
for (mode = '1'; mode <= '4'; ++mode) {
int unlimited;
for (unlimited = 0; unlimited < 2; unlimited++) {
wchar_t buf[BUFSIZE];
const char *src;
mbstate_t temp_state;
{
size_t i;
for (i = 0; i < BUFSIZE; i++)
buf[i] = (wchar_t) 0xBADFACE;
}
switch (mode) {
case '1':
/* Locale encoding is ISO-8859-1 or ISO-8859-15. */
printf("ISO8859-1 ...\n");
{
char input[] = "B\374\337er"; /* "Büßer" */
memset(&state, '\0', sizeof(mbstate_t));
if (setlocale (LC_ALL, "en_US.ISO8859-1") == NULL) {
fprintf(stderr,
"unable to set ISO8859-1 locale, skipping\n");
break;
}
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc(&wc, input, 1, &state);
assert(ret == 1);
assert(wc == 'B');
assert(mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc(&wc, input + 1, 1, &state);
assert(ret == 1);
assert(wctob (wc) == (unsigned char) '\374');
assert(mbsinit (&state));
input[1] = '\0';
src = input + 2;
temp_state = state;
ret = mbsnrtowcs(NULL, &src, 4, unlimited ? BUFSIZE : 1,
&temp_state);
assert(ret == 3);
assert(src == input + 2);
assert(mbsinit (&state));
src = input + 2;
ret = mbsnrtowcs(buf, &src, 4, unlimited ? BUFSIZE : 1,
&state);
assert(ret == (unlimited ? 3 : 1));
assert(src == (unlimited ? NULL : input + 3));
assert(wctob (buf[0]) == (unsigned char) '\337');
if (unlimited) {
assert(buf[1] == 'e');
assert(buf[2] == 'r');
assert(buf[3] == 0);
assert(buf[4] == (wchar_t) 0xBADFACE);
} else
assert(buf[1] == (wchar_t) 0xBADFACE);
assert(mbsinit (&state));
}
break;
case '2':
/* Locale encoding is UTF-8. */
printf("UTF-8 ...\n");
{
char input[] = "B\303\274\303\237er"; /* "Büßer" */
memset(&state, '\0', sizeof(mbstate_t));
if (setlocale (LC_ALL, "en_US.UTF-8") == NULL) {
fprintf(stderr,
"unable to set UTF-8 locale, skipping\n");
break;
}
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc(&wc, input, 1, &state);
assert(ret == 1);
assert(wc == 'B');
assert(mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc(&wc, input + 1, 1, &state);
assert(ret == (size_t)(-2));
assert(wc == (wchar_t) 0xBADFACE);
assert(!mbsinit (&state));
input[1] = '\0';
// Copying mbstate_t doesn't really copy the ICU-converter's state, so this
// doesn't work on Haiku.
#ifndef __HAIKU__
src = input + 2;
temp_state = state;
ret = mbsnrtowcs(NULL, &src, 6, unlimited ? BUFSIZE : 2,
&temp_state);
assert(ret == 4);
assert(src == input + 2);
assert(!mbsinit (&state));
#endif
src = input + 2;
ret = mbsnrtowcs(buf, &src, 6, unlimited ? BUFSIZE : 2,
&state);
assert(ret == (unlimited ? 4 : 2));
assert(src == (unlimited ? NULL : input + 5));
assert(wctob (buf[0]) == EOF);
assert(wctob (buf[1]) == EOF);
if (unlimited) {
assert(buf[2] == 'e');
assert(buf[3] == 'r');
assert(buf[4] == 0);
assert(buf[5] == (wchar_t) 0xBADFACE);
} else
assert(buf[2] == (wchar_t) 0xBADFACE);
assert(mbsinit (&state));
}
break;
case '3':
/* Locale encoding is EUC-JP. */
printf("EUC-JP ...\n");
{
char input[] = "<\306\374\313\334\270\354>"; /* "<日本語>" */
memset(&state, '\0', sizeof(mbstate_t));
if (setlocale (LC_ALL, "en_US.EUC-JP") == NULL) {
fprintf(stderr,
"unable to set EUC-JP locale, skipping\n");
break;
}
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc(&wc, input, 1, &state);
assert(ret == 1);
assert(wc == '<');
assert(mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc(&wc, input + 1, 2, &state);
assert(ret == 2);
assert(wctob (wc) == EOF);
assert(mbsinit (&state));
input[1] = '\0';
input[2] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc(&wc, input + 3, 1, &state);
assert(ret == (size_t)(-2));
assert(wc == (wchar_t) 0xBADFACE);
assert(!mbsinit (&state));
input[3] = '\0';
// Copying mbstate_t doesn't really copy the ICU-converter's state, so this
// doesn't work on Haiku.
#ifndef __HAIKU__
src = input + 4;
temp_state = state;
ret = mbsnrtowcs(NULL, &src, 5, unlimited ? BUFSIZE : 2,
&temp_state);
assert(ret == 3);
assert(src == input + 4);
assert(!mbsinit (&state));
#endif
src = input + 4;
ret = mbsnrtowcs(buf, &src, 5, unlimited ? BUFSIZE : 2,
&state);
assert(ret == (unlimited ? 3 : 2));
assert(src == (unlimited ? NULL : input + 7));
assert(wctob (buf[0]) == EOF);
assert(wctob (buf[1]) == EOF);
if (unlimited) {
assert(buf[2] == '>');
assert(buf[3] == 0);
assert(buf[4] == (wchar_t) 0xBADFACE);
} else
assert(buf[2] == (wchar_t) 0xBADFACE);
assert(mbsinit (&state));
}
break;
case '4':
/* Locale encoding is GB18030. */
printf("GB18030 ...\n");
{
char input[] = "B\250\271\201\060\211\070er"; /* "Büßer" */
memset(&state, '\0', sizeof(mbstate_t));
if (setlocale (LC_ALL, "en_US.GB18030") == NULL) {
fprintf(stderr,
"unable to set GB18030 locale, skipping\n");
break;
}
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc(&wc, input, 1, &state);
assert(ret == 1);
assert(wc == 'B');
assert(mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc(&wc, input + 1, 1, &state);
assert(ret == (size_t)(-2));
assert(wc == (wchar_t) 0xBADFACE);
assert(!mbsinit (&state));
input[1] = '\0';
// Copying mbstate_t doesn't really copy the ICU-converter's state, so this
// doesn't work on Haiku.
#ifndef __HAIKU__
src = input + 2;
temp_state = state;
ret = mbsnrtowcs(NULL, &src, 8, unlimited ? BUFSIZE : 2,
&temp_state);
assert(ret == 4);
assert(src == input + 2);
assert(!mbsinit (&state));
#endif
src = input + 2;
ret = mbsnrtowcs(buf, &src, 8, unlimited ? BUFSIZE : 2,
&state);
assert(ret == (unlimited ? 4 : 2));
assert(src == (unlimited ? NULL : input + 7));
assert(wctob (buf[0]) == EOF);
assert(wctob (buf[1]) == EOF);
if (unlimited) {
assert(buf[2] == 'e');
assert(buf[3] == 'r');
assert(buf[4] == 0);
assert(buf[5] == (wchar_t) 0xBADFACE);
} else
assert(buf[2] == (wchar_t) 0xBADFACE);
assert(mbsinit (&state));
}
break;
default:
return 1;
}
}
}
return 0;
}
static int
test_one_locale (const char *name, int codepage)
{
mbstate_t state;
wchar_t wc;
size_t ret;
# if 1
/* Portable code to set the locale. */
{
char name_with_codepage[1024];
sprintf (name_with_codepage, "%s.%d", name, codepage);
/* Set the locale. */
if (setlocale (LC_ALL, name_with_codepage) == NULL)
return 77;
}
# else
/* Hacky way to set a locale.codepage combination that setlocale() refuses
to set. */
{
/* Codepage of the current locale, set with setlocale().
Not necessarily the same as GetACP(). */
extern __declspec(dllimport) unsigned int __lc_codepage;
/* Set the locale. */
if (setlocale (LC_ALL, name) == NULL)
return 77;
/* Clobber the codepage and MB_CUR_MAX, both set by setlocale(). */
__lc_codepage = codepage;
switch (codepage)
{
case 1252:
case 1256:
MB_CUR_MAX = 1;
break;
case 932:
case 950:
case 936:
MB_CUR_MAX = 2;
break;
case 54936:
case 65001:
MB_CUR_MAX = 4;
break;
}
/* Test whether the codepage is really available. */
memset (&state, '\0', sizeof (mbstate_t));
if (mbrtowc (&wc, " ", 1, &state) == (size_t)(-1))
return 77;
}
# endif
/* Test zero-length input. */
{
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "x", 0, &state);
/* gnulib's implementation returns (size_t)(-2).
The AIX 5.1 implementation returns (size_t)(-1).
glibc's implementation returns 0. */
ASSERT (ret == (size_t)(-2) || ret == (size_t)(-1) || ret == 0);
ASSERT (mbsinit (&state));
}
/* Test NUL byte input. */
{
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "", 1, &state);
ASSERT (ret == 0);
ASSERT (wc == 0);
ASSERT (mbsinit (&state));
ret = mbrtowc (NULL, "", 1, &state);
ASSERT (ret == 0);
ASSERT (mbsinit (&state));
}
/* Test single-byte input. */
{
int c;
char buf[1];
memset (&state, '\0', sizeof (mbstate_t));
for (c = 0; c < 0x100; c++)
switch (c)
{
case '\t': case '\v': case '\f':
case ' ': case '!': case '"': case '#': case '%':
case '&': case '\'': case '(': case ')': case '*':
case '+': case ',': case '-': case '.': case '/':
case '0': case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
case ':': case ';': case '<': case '=': case '>':
case '?':
case 'A': case 'B': case 'C': case 'D': case 'E':
case 'F': case 'G': case 'H': case 'I': case 'J':
case 'K': case 'L': case 'M': case 'N': case 'O':
case 'P': case 'Q': case 'R': case 'S': case 'T':
case 'U': case 'V': case 'W': case 'X': case 'Y':
case 'Z':
case '[': case '\\': case ']': case '^': case '_':
case 'a': case 'b': case 'c': case 'd': case 'e':
case 'f': case 'g': case 'h': case 'i': case 'j':
case 'k': case 'l': case 'm': case 'n': case 'o':
case 'p': case 'q': case 'r': case 's': case 't':
case 'u': case 'v': case 'w': case 'x': case 'y':
case 'z': case '{': case '|': case '}': case '~':
/* c is in the ISO C "basic character set". */
buf[0] = c;
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, buf, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == c);
ASSERT (mbsinit (&state));
ret = mbrtowc (NULL, buf, 1, &state);
ASSERT (ret == 1);
ASSERT (mbsinit (&state));
break;
}
}
/* Test special calling convention, passing a NULL pointer. */
{
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, NULL, 5, &state);
ASSERT (ret == 0);
ASSERT (wc == (wchar_t) 0xBADFACE);
ASSERT (mbsinit (&state));
}
switch (codepage)
{
case 1252:
/* Locale encoding is CP1252, an extension of ISO-8859-1. */
{
char input[] = "B\374\337er"; /* "Büßer" */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'B');
ASSERT (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 1, &state);
ASSERT (ret == 1);
ASSERT (wctob (wc) == (unsigned char) '\374');
ASSERT (wc == 0x00FC);
ASSERT (mbsinit (&state));
input[1] = '\0';
/* Test support of NULL first argument. */
ret = mbrtowc (NULL, input + 2, 3, &state);
ASSERT (ret == 1);
ASSERT (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 2, 3, &state);
ASSERT (ret == 1);
ASSERT (wctob (wc) == (unsigned char) '\337');
ASSERT (wc == 0x00DF);
ASSERT (mbsinit (&state));
input[2] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 3, 2, &state);
ASSERT (ret == 1);
ASSERT (wc == 'e');
ASSERT (mbsinit (&state));
input[3] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 4, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'r');
ASSERT (mbsinit (&state));
}
return 0;
case 1256:
/* Locale encoding is CP1256, not the same as ISO-8859-6. */
{
char input[] = "x\302\341\346y"; /* "xآلوy" */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'x');
ASSERT (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 1, &state);
ASSERT (ret == 1);
ASSERT (wctob (wc) == (unsigned char) '\302');
ASSERT (wc == 0x0622);
ASSERT (mbsinit (&state));
input[1] = '\0';
/* Test support of NULL first argument. */
ret = mbrtowc (NULL, input + 2, 3, &state);
ASSERT (ret == 1);
ASSERT (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 2, 3, &state);
ASSERT (ret == 1);
ASSERT (wctob (wc) == (unsigned char) '\341');
ASSERT (wc == 0x0644);
ASSERT (mbsinit (&state));
input[2] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 3, 2, &state);
ASSERT (ret == 1);
ASSERT (wctob (wc) == (unsigned char) '\346');
ASSERT (wc == 0x0648);
ASSERT (mbsinit (&state));
input[3] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 4, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'y');
ASSERT (mbsinit (&state));
}
return 0;
case 932:
/* Locale encoding is CP932, similar to Shift_JIS. */
{
char input[] = "<\223\372\226\173\214\352>"; /* "<日本語>" */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == '<');
ASSERT (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 2, &state);
ASSERT (ret == 2);
ASSERT (wctob (wc) == EOF);
ASSERT (wc == 0x65E5);
ASSERT (mbsinit (&state));
input[1] = '\0';
input[2] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 3, 1, &state);
ASSERT (ret == (size_t)(-2));
ASSERT (wc == (wchar_t) 0xBADFACE);
ASSERT (!mbsinit (&state));
input[3] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 4, 4, &state);
ASSERT (ret == 1);
ASSERT (wctob (wc) == EOF);
ASSERT (wc == 0x672C);
ASSERT (mbsinit (&state));
input[4] = '\0';
/* Test support of NULL first argument. */
ret = mbrtowc (NULL, input + 5, 3, &state);
ASSERT (ret == 2);
ASSERT (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 5, 3, &state);
ASSERT (ret == 2);
ASSERT (wctob (wc) == EOF);
ASSERT (wc == 0x8A9E);
ASSERT (mbsinit (&state));
input[5] = '\0';
input[6] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 7, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == '>');
ASSERT (mbsinit (&state));
/* Test some invalid input. */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\377", 1, &state); /* 0xFF */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\225\377", 2, &state); /* 0x95 0xFF */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
}
return 0;
case 950:
/* Locale encoding is CP950, similar to Big5. */
{
char input[] = "<\244\351\245\273\273\171>"; /* "<日本語>" */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == '<');
ASSERT (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 2, &state);
ASSERT (ret == 2);
ASSERT (wctob (wc) == EOF);
ASSERT (wc == 0x65E5);
ASSERT (mbsinit (&state));
input[1] = '\0';
input[2] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 3, 1, &state);
ASSERT (ret == (size_t)(-2));
ASSERT (wc == (wchar_t) 0xBADFACE);
ASSERT (!mbsinit (&state));
input[3] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 4, 4, &state);
ASSERT (ret == 1);
ASSERT (wctob (wc) == EOF);
ASSERT (wc == 0x672C);
ASSERT (mbsinit (&state));
input[4] = '\0';
/* Test support of NULL first argument. */
ret = mbrtowc (NULL, input + 5, 3, &state);
ASSERT (ret == 2);
ASSERT (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 5, 3, &state);
ASSERT (ret == 2);
ASSERT (wctob (wc) == EOF);
ASSERT (wc == 0x8A9E);
ASSERT (mbsinit (&state));
input[5] = '\0';
input[6] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 7, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == '>');
ASSERT (mbsinit (&state));
/* Test some invalid input. */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\377", 1, &state); /* 0xFF */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\225\377", 2, &state); /* 0x95 0xFF */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
}
return 0;
case 936:
/* Locale encoding is CP936 = GBK, an extension of GB2312. */
{
char input[] = "<\310\325\261\276\325\132>"; /* "<日本語>" */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == '<');
ASSERT (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 2, &state);
ASSERT (ret == 2);
ASSERT (wctob (wc) == EOF);
ASSERT (wc == 0x65E5);
ASSERT (mbsinit (&state));
input[1] = '\0';
input[2] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 3, 1, &state);
ASSERT (ret == (size_t)(-2));
ASSERT (wc == (wchar_t) 0xBADFACE);
ASSERT (!mbsinit (&state));
input[3] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 4, 4, &state);
ASSERT (ret == 1);
ASSERT (wctob (wc) == EOF);
ASSERT (wc == 0x672C);
ASSERT (mbsinit (&state));
input[4] = '\0';
/* Test support of NULL first argument. */
ret = mbrtowc (NULL, input + 5, 3, &state);
ASSERT (ret == 2);
ASSERT (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 5, 3, &state);
ASSERT (ret == 2);
ASSERT (wctob (wc) == EOF);
ASSERT (wc == 0x8A9E);
ASSERT (mbsinit (&state));
input[5] = '\0';
input[6] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 7, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == '>');
ASSERT (mbsinit (&state));
/* Test some invalid input. */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\377", 1, &state); /* 0xFF */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\225\377", 2, &state); /* 0x95 0xFF */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
}
return 0;
case 54936:
/* Locale encoding is CP54936 = GB18030. */
{
char input[] = "B\250\271\201\060\211\070er"; /* "Büßer" */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'B');
ASSERT (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 1, &state);
ASSERT (ret == (size_t)(-2));
ASSERT (wc == (wchar_t) 0xBADFACE);
ASSERT (!mbsinit (&state));
input[1] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 2, 7, &state);
ASSERT (ret == 1);
ASSERT (wctob (wc) == EOF);
ASSERT (wc == 0x00FC);
ASSERT (mbsinit (&state));
input[2] = '\0';
/* Test support of NULL first argument. */
ret = mbrtowc (NULL, input + 3, 6, &state);
ASSERT (ret == 4);
ASSERT (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 3, 6, &state);
ASSERT (ret == 4);
ASSERT (wctob (wc) == EOF);
ASSERT (wc == 0x00DF);
ASSERT (mbsinit (&state));
input[3] = '\0';
input[4] = '\0';
input[5] = '\0';
input[6] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 7, 2, &state);
ASSERT (ret == 1);
ASSERT (wc == 'e');
ASSERT (mbsinit (&state));
input[5] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 8, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'r');
ASSERT (mbsinit (&state));
/* Test some invalid input. */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\377", 1, &state); /* 0xFF */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\225\377", 2, &state); /* 0x95 0xFF */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\201\045", 2, &state); /* 0x81 0x25 */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\201\060\377", 3, &state); /* 0x81 0x30 0xFF */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\201\060\377\064", 4, &state); /* 0x81 0x30 0xFF 0x34 */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\201\060\211\072", 4, &state); /* 0x81 0x30 0x89 0x3A */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
}
return 0;
case 65001:
/* Locale encoding is CP65001 = UTF-8. */
{
char input[] = "B\303\274\303\237er"; /* "Büßer" */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'B');
ASSERT (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 1, &state);
ASSERT (ret == (size_t)(-2));
ASSERT (wc == (wchar_t) 0xBADFACE);
ASSERT (!mbsinit (&state));
input[1] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 2, 5, &state);
ASSERT (ret == 1);
ASSERT (wctob (wc) == EOF);
ASSERT (wc == 0x00FC);
ASSERT (mbsinit (&state));
input[2] = '\0';
/* Test support of NULL first argument. */
ret = mbrtowc (NULL, input + 3, 4, &state);
ASSERT (ret == 2);
ASSERT (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 3, 4, &state);
ASSERT (ret == 2);
ASSERT (wctob (wc) == EOF);
ASSERT (wc == 0x00DF);
ASSERT (mbsinit (&state));
input[3] = '\0';
input[4] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 5, 2, &state);
ASSERT (ret == 1);
ASSERT (wc == 'e');
ASSERT (mbsinit (&state));
input[5] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 6, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'r');
ASSERT (mbsinit (&state));
/* Test some invalid input. */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\377", 1, &state); /* 0xFF */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\303\300", 2, &state); /* 0xC3 0xC0 */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\343\300", 2, &state); /* 0xE3 0xC0 */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\343\300\200", 3, &state); /* 0xE3 0xC0 0x80 */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\343\200\300", 3, &state); /* 0xE3 0x80 0xC0 */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\363\300", 2, &state); /* 0xF3 0xC0 */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\363\300\200\200", 4, &state); /* 0xF3 0xC0 0x80 0x80 */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\363\200\300", 3, &state); /* 0xF3 0x80 0xC0 */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\363\200\300\200", 4, &state); /* 0xF3 0x80 0xC0 0x80 */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "\363\200\200\300", 4, &state); /* 0xF3 0x80 0x80 0xC0 */
ASSERT (ret == (size_t)-1);
ASSERT (errno == EILSEQ);
}
return 0;
default:
return 1;
}
}
static size_t
multibyte_qmark_chars(char *buf, size_t len)
{
if (MB_CUR_MAX <= 1)
{
return unibyte_qmark_chars(buf, len);
}
else
{
char const *p = buf;
char const *plimit = buf + len;
char *q = buf;
while (p < plimit)
switch (*p)
{
case ' ': case '!': case '"': case '#': case '%':
case '&': case '\'': case '(': case ')': case '*':
case '+': case ',': case '-': case '.': case '/':
case '0': case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
case ':': case ';': case '<': case '=': case '>':
case '?':
case 'A': case 'B': case 'C': case 'D': case 'E':
case 'F': case 'G': case 'H': case 'I': case 'J':
case 'K': case 'L': case 'M': case 'N': case 'O':
case 'P': case 'Q': case 'R': case 'S': case 'T':
case 'U': case 'V': case 'W': case 'X': case 'Y':
case 'Z':
case '[': case '\\': case ']': case '^': case '_':
case 'a': case 'b': case 'c': case 'd': case 'e':
case 'f': case 'g': case 'h': case 'i': case 'j':
case 'k': case 'l': case 'm': case 'n': case 'o':
case 'p': case 'q': case 'r': case 's': case 't':
case 'u': case 'v': case 'w': case 'x': case 'y':
case 'z': case '{': case '|': case '}': case '~':
/* These characters are printable ASCII characters. */
*q++ = *p++;
break;
default:
/* If we have a multibyte sequence, copy it until we
reach its end, replacing each non-printable multibyte
character with a single question mark. */
{
mbstate_t mbstate;
memset (&mbstate, 0, sizeof mbstate);
do
{
wchar_t wc;
size_t bytes;
int w;
bytes = mbrtowc (&wc, p, plimit - p, &mbstate);
if (bytes == (size_t) -1)
{
/* An invalid multibyte sequence was
encountered. Skip one input byte, and
put a question mark. */
p++;
*q++ = '?';
break;
}
if (bytes == (size_t) -2)
{
/* An incomplete multibyte character
at the end. Replace it entirely with
a question mark. */
p = plimit;
*q++ = '?';
break;
}
if (bytes == 0)
/* A null wide character was encountered. */
bytes = 1;
w = wcwidth (wc);
if (w >= 0)
{
/* A printable multibyte character.
Keep it. */
for (; bytes > 0; --bytes)
*q++ = *p++;
}
else
{
/* An unprintable multibyte character.
Replace it entirely with a question
mark. */
p += bytes;
*q++ = '?';
}
}
while (! mbsinit (&mbstate));
}
break;
}
/* The buffer may have shrunk. */
len = q - buf;
return len;
}
}
int main()
{
mbstate_t mb = {0};
size_t s = 0;
tm tm = {0};
wint_t w = 0;
::FILE* fp = 0;
__darwin_va_list va;
char* ns = 0;
wchar_t* ws = 0;
static_assert((std::is_same<decltype(fwprintf(fp, L"")), int>::value), "");
static_assert((std::is_same<decltype(fwscanf(fp, L"")), int>::value), "");
static_assert((std::is_same<decltype(swprintf(ws, s, L"")), int>::value), "");
static_assert((std::is_same<decltype(swscanf(L"", L"")), int>::value), "");
static_assert((std::is_same<decltype(vfwprintf(fp, L"", va)), int>::value), "");
static_assert((std::is_same<decltype(vfwscanf(fp, L"", va)), int>::value), "");
static_assert((std::is_same<decltype(vswprintf(ws, s, L"", va)), int>::value), "");
static_assert((std::is_same<decltype(vswscanf(L"", L"", va)), int>::value), "");
static_assert((std::is_same<decltype(vwprintf(L"", va)), int>::value), "");
static_assert((std::is_same<decltype(vwscanf(L"", va)), int>::value), "");
static_assert((std::is_same<decltype(wprintf(L"")), int>::value), "");
static_assert((std::is_same<decltype(wscanf(L"")), int>::value), "");
static_assert((std::is_same<decltype(fgetwc(fp)), wint_t>::value), "");
static_assert((std::is_same<decltype(fgetws(ws, 0, fp)), wchar_t*>::value), "");
static_assert((std::is_same<decltype(fputwc(L' ', fp)), wint_t>::value), "");
static_assert((std::is_same<decltype(fputws(L"", fp)), int>::value), "");
static_assert((std::is_same<decltype(fwide(fp, 0)), int>::value), "");
static_assert((std::is_same<decltype(getwc(fp)), wint_t>::value), "");
static_assert((std::is_same<decltype(getwchar()), wint_t>::value), "");
static_assert((std::is_same<decltype(putwc(L' ', fp)), wint_t>::value), "");
static_assert((std::is_same<decltype(putwchar(L' ')), wint_t>::value), "");
static_assert((std::is_same<decltype(ungetwc(L' ', fp)), wint_t>::value), "");
static_assert((std::is_same<decltype(wcstod(L"", (wchar_t**)0)), double>::value), "");
static_assert((std::is_same<decltype(wcstof(L"", (wchar_t**)0)), float>::value), "");
static_assert((std::is_same<decltype(wcstold(L"", (wchar_t**)0)), long double>::value), "");
static_assert((std::is_same<decltype(wcstol(L"", (wchar_t**)0, 0)), long>::value), "");
static_assert((std::is_same<decltype(wcstoll(L"", (wchar_t**)0, 0)), long long>::value), "");
static_assert((std::is_same<decltype(wcstoul(L"", (wchar_t**)0, 0)), unsigned long>::value), "");
static_assert((std::is_same<decltype(wcstoull(L"", (wchar_t**)0, 0)), unsigned long long>::value), "");
static_assert((std::is_same<decltype(wcscpy(ws, L"")), wchar_t*>::value), "");
static_assert((std::is_same<decltype(wcsncpy(ws, L"", s)), wchar_t*>::value), "");
static_assert((std::is_same<decltype(wcscat(ws, L"")), wchar_t*>::value), "");
static_assert((std::is_same<decltype(wcsncat(ws, L"", s)), wchar_t*>::value), "");
static_assert((std::is_same<decltype(wcscmp(L"", L"")), int>::value), "");
static_assert((std::is_same<decltype(wcscoll(L"", L"")), int>::value), "");
static_assert((std::is_same<decltype(wcsncmp(L"", L"", s)), int>::value), "");
static_assert((std::is_same<decltype(wcsxfrm(ws, L"", s)), size_t>::value), "");
static_assert((std::is_same<decltype(wcschr((wchar_t*)0, L' ')), wchar_t*>::value), "");
static_assert((std::is_same<decltype(wcscspn(L"", L"")), size_t>::value), "");
static_assert((std::is_same<decltype(wcslen(L"")), size_t>::value), "");
static_assert((std::is_same<decltype(wcspbrk((wchar_t*)0, L"")), wchar_t*>::value), "");
static_assert((std::is_same<decltype(wcsrchr((wchar_t*)0, L' ')), wchar_t*>::value), "");
static_assert((std::is_same<decltype(wcsspn(L"", L"")), size_t>::value), "");
static_assert((std::is_same<decltype(wcsstr((wchar_t*)0, L"")), wchar_t*>::value), "");
static_assert((std::is_same<decltype(wcstok(ws, L"", (wchar_t**)0)), wchar_t*>::value), "");
static_assert((std::is_same<decltype(wmemchr((wchar_t*)0, L' ', s)), wchar_t*>::value), "");
static_assert((std::is_same<decltype(wmemcmp(L"", L"", s)), int>::value), "");
static_assert((std::is_same<decltype(wmemcpy(ws, L"", s)), wchar_t*>::value), "");
static_assert((std::is_same<decltype(wmemmove(ws, L"", s)), wchar_t*>::value), "");
static_assert((std::is_same<decltype(wmemset(ws, L' ', s)), wchar_t*>::value), "");
static_assert((std::is_same<decltype(wcsftime(ws, s, L"", &tm)), size_t>::value), "");
static_assert((std::is_same<decltype(btowc(0)), wint_t>::value), "");
static_assert((std::is_same<decltype(wctob(w)), int>::value), "");
static_assert((std::is_same<decltype(mbsinit(&mb)), int>::value), "");
static_assert((std::is_same<decltype(mbrlen("", s, &mb)), size_t>::value), "");
static_assert((std::is_same<decltype(mbrtowc(ws, "", s, &mb)), size_t>::value), "");
static_assert((std::is_same<decltype(wcrtomb(ns, L' ', &mb)), size_t>::value), "");
static_assert((std::is_same<decltype(mbsrtowcs(ws, (const char**)0, s, &mb)), size_t>::value), "");
static_assert((std::is_same<decltype(wcsrtombs(ns, (const wchar_t**)0, s, &mb)), size_t>::value), "");
}
static int
processStream (FILE *inputStream, const char *inputName) {
mbstate_t inputState;
mbstate_t outputState;
memset(&inputState, 0, sizeof(inputState));
memset(&outputState, 0, sizeof(outputState));
while (!feof(inputStream)) {
char inputBuffer[0X1000];
size_t inputCount = fread(inputBuffer, 1, sizeof(inputBuffer), inputStream);
if (ferror(inputStream)) goto inputError;
if (!inputCount) break;
{
char *byte = inputBuffer;
while (inputCount) {
wchar_t character;
{
size_t result = mbrtowc(&character, byte, inputCount, &inputState);
if (result == (size_t)-2) break;
if (result == (size_t)-1) goto inputError;
if (!result) result = 1;
byte += result;
inputCount -= result;
}
if (!iswcntrl(character)) {
unsigned char dots = toDots(character);
if (dots || !iswspace(character)) {
if (opt_sixDots) dots &= ~(BRL_DOT_7 | BRL_DOT_8);
character = toCharacter(dots);
}
}
if (!writeCharacter(&character, &outputState)) goto outputError;
}
}
}
if (!writeCharacter(NULL, &outputState)) goto outputError;
fflush(outputStream);
if (ferror(outputStream)) goto outputError;
if (!mbsinit(&inputState)) {
#ifdef EILSEQ
errno = EILSEQ;
#else /* EILSEQ */
errno = EINVAL;
#endif /* EILSEQ */
goto inputError;
}
return 1;
inputError:
logMessage(LOG_ERR, "input error: %s: %s", inputName, strerror(errno));
return 0;
outputError:
logMessage(LOG_ERR, "output error: %s: %s", outputName, strerror(errno));
return 0;
}
/**
* ntfs_ucstombs - convert a little endian Unicode string to a multibyte string
* @ins: input Unicode string buffer
* @ins_len: length of input string in Unicode characters
* @outs: on return contains the (allocated) output multibyte string
* @outs_len: length of output buffer in bytes
*
* Convert the input little endian, 2-byte Unicode string @ins, of length
* @ins_len into the multibyte string format dictated by the current locale.
*
* If *@outs is NULL, the function allocates the string and the caller is
* responsible for calling free(*@outs); when finished with it.
*
* On success the function returns the number of bytes written to the output
* string *@outs (>= 0), not counting the terminating NULL byte. If the output
* string buffer was allocated, *@outs is set to it.
*
* On error, -1 is returned, and errno is set to the error code. The following
* error codes can be expected:
* EINVAL Invalid arguments (e.g. @ins or @outs is NULL).
* EILSEQ The input string cannot be represented as a multibyte
* sequence according to the current locale.
* ENAMETOOLONG Destination buffer is too small for input string.
* ENOMEM Not enough memory to allocate destination buffer.
*/
int ntfs_ucstombs(const ntfschar *ins, const int ins_len, char **outs,
int outs_len)
{
char *mbs;
wchar_t wc;
int i, o, mbs_len;
int cnt = 0;
#ifdef HAVE_MBSINIT
mbstate_t mbstate;
#endif
if (!ins || !outs) {
errno = EINVAL;
return -1;
}
mbs = *outs;
mbs_len = outs_len;
if (mbs && !mbs_len) {
errno = ENAMETOOLONG;
return -1;
}
if (!mbs) {
mbs_len = (ins_len + 1) * MB_CUR_MAX;
mbs = (char*)malloc(mbs_len);
if (!mbs)
return -1;
}
#ifdef HAVE_MBSINIT
memset(&mbstate, 0, sizeof(mbstate));
#else
wctomb(NULL, 0);
#endif
for (i = o = 0; i < ins_len; i++) {
/* Reallocate memory if necessary or abort. */
if ((int)(o + MB_CUR_MAX) > mbs_len) {
char *tc;
if (mbs == *outs) {
errno = ENAMETOOLONG;
return -1;
}
tc = (char*)malloc((mbs_len + 64) & ~63);
if (!tc)
goto err_out;
memcpy(tc, mbs, mbs_len);
mbs_len = (mbs_len + 64) & ~63;
free(mbs);
mbs = tc;
}
/* Convert the LE Unicode character to a CPU wide character. */
wc = (wchar_t)le16_to_cpu(ins[i]);
if (!wc)
break;
/* Convert the CPU endian wide character to multibyte. */
#ifdef HAVE_MBSINIT
cnt = wcrtomb(mbs + o, wc, &mbstate);
#else
cnt = wctomb(mbs + o, wc);
#endif
if (cnt == -1)
goto err_out;
if (cnt <= 0) {
ntfs_log_debug("Eeek. cnt <= 0, cnt = %i\n", cnt);
errno = EINVAL;
goto err_out;
}
o += cnt;
}
#ifdef HAVE_MBSINIT
/* Make sure we are back in the initial state. */
if (!mbsinit(&mbstate)) {
ntfs_log_debug("Eeek. mbstate not in initial state!\n");
errno = EILSEQ;
goto err_out;
}
#endif
/* Now write the NULL character. */
mbs[o] = '\0';
if (*outs != mbs)
*outs = mbs;
return o;
err_out:
if (mbs != *outs) {
int eo = errno;
free(mbs);
errno = eo;
}
return -1;
}
size_t wcsnrtombs(char* dst, const wchar_t** src, size_t nwc, size_t len, mbstate_t* ps) {
static mbstate_t __private_state;
mbstate_t* state = (ps == NULL) ? &__private_state : ps;
if (!mbsinit(state)) {
return reset_and_return_illegal(EILSEQ, state);
}
char buf[MB_LEN_MAX];
size_t i, o, r;
if (dst == NULL) {
for (i = o = 0; i < nwc; i++, o += r) {
wchar_t wc = (*src)[i];
if (static_cast<uint32_t>(wc) < 0x80) {
// Fast path for plain ASCII characters.
if (wc == 0) {
return o;
}
r = 1;
} else {
r = wcrtomb(buf, wc, state);
if (r == ERR_ILLEGAL_SEQUENCE) {
return r;
}
}
}
return o;
}
for (i = o = 0; i < nwc && o < len; i++, o += r) {
wchar_t wc = (*src)[i];
if (static_cast<uint32_t>(wc) < 0x80) {
// Fast path for plain ASCII characters.
dst[o] = wc;
if (wc == 0) {
*src = NULL;
return o;
}
r = 1;
} else if (len - o >= sizeof(buf)) {
// Enough space to translate in-place.
r = wcrtomb(dst + o, wc, state);
if (r == ERR_ILLEGAL_SEQUENCE) {
*src += i;
return r;
}
} else {
// May not be enough space; use temp buffer.
r = wcrtomb(buf, wc, state);
if (r == ERR_ILLEGAL_SEQUENCE) {
*src += i;
return r;
}
if (r > len - o) {
break;
}
memcpy(dst + o, buf, r);
}
}
*src += i;
return o;
}
size_t wcrtomb(char* s, wchar_t wc, mbstate_t* ps) {
static mbstate_t __private_state;
mbstate_t* state = (ps == NULL) ? &__private_state : ps;
if (s == NULL) {
// Equivalent to wcrtomb(buf, L'\0', ps).
return reset_and_return(1, state);
}
// POSIX states that if wc is a null wide character, a null byte shall be
// stored, preceded by any shift sequence needed to restore the initial shift
// state. Since shift states are not supported, only the null byte is stored.
if (wc == L'\0') {
*s = '\0';
reset_and_return(1, state);
}
if (!mbsinit(state)) {
return reset_and_return_illegal(EILSEQ, state);
}
if ((wc & ~0x7f) == 0) {
// Fast path for plain ASCII characters.
*s = wc;
return 1;
}
// Determine the number of octets needed to represent this character.
// We always output the shortest sequence possible. Also specify the
// first few bits of the first octet, which contains the information
// about the sequence length.
uint8_t lead;
size_t length;
if ((wc & ~0x7f) == 0) {
lead = 0;
length = 1;
} else if ((wc & ~0x7ff) == 0) {
lead = 0xc0;
length = 2;
} else if ((wc & ~0xffff) == 0) {
lead = 0xe0;
length = 3;
} else if ((wc & ~0x1fffff) == 0) {
lead = 0xf0;
length = 4;
} else {
errno = EILSEQ;
return ERR_ILLEGAL_SEQUENCE;
}
// Output the octets representing the character in chunks
// of 6 bits, least significant last. The first octet is
// a special case because it contains the sequence length
// information.
for (size_t i = length - 1; i > 0; i--) {
s[i] = (wc & 0x3f) | 0x80;
wc >>= 6;
}
*s = (wc & 0xff) | lead;
return length;
}
int
main (int argc, char *argv[])
{
mbstate_t state;
wchar_t wc;
size_t ret;
/* configure should already have checked that the locale is supported. */
if (setlocale (LC_ALL, "") == NULL)
return 1;
/* Test zero-length input. */
{
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "x", 0, &state);
ASSERT (ret == (size_t)(-2));
ASSERT (mbsinit (&state));
}
/* Test NUL byte input. */
{
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "", 1, &state);
ASSERT (ret == 0);
ASSERT (wc == 0);
ASSERT (mbsinit (&state));
ret = mbrtowc (NULL, "", 1, &state);
ASSERT (ret == 0);
ASSERT (mbsinit (&state));
}
/* Test single-byte input. */
{
int c;
char buf[1];
memset (&state, '\0', sizeof (mbstate_t));
for (c = 0; c < 0x100; c++)
switch (c)
{
case '\t': case '\v': case '\f':
case ' ': case '!': case '"': case '#': case '%':
case '&': case '\'': case '(': case ')': case '*':
case '+': case ',': case '-': case '.': case '/':
case '0': case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
case ':': case ';': case '<': case '=': case '>':
case '?':
case 'A': case 'B': case 'C': case 'D': case 'E':
case 'F': case 'G': case 'H': case 'I': case 'J':
case 'K': case 'L': case 'M': case 'N': case 'O':
case 'P': case 'Q': case 'R': case 'S': case 'T':
case 'U': case 'V': case 'W': case 'X': case 'Y':
case 'Z':
case '[': case '\\': case ']': case '^': case '_':
case 'a': case 'b': case 'c': case 'd': case 'e':
case 'f': case 'g': case 'h': case 'i': case 'j':
case 'k': case 'l': case 'm': case 'n': case 'o':
case 'p': case 'q': case 'r': case 's': case 't':
case 'u': case 'v': case 'w': case 'x': case 'y':
case 'z': case '{': case '|': case '}': case '~':
/* c is in the ISO C "basic character set". */
buf[0] = c;
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, buf, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == c);
ASSERT (mbsinit (&state));
ret = mbrtowc (NULL, buf, 1, &state);
ASSERT (ret == 1);
ASSERT (mbsinit (&state));
break;
}
}
/* Test special calling convention, passing a NULL pointer. */
{
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, NULL, 5, &state);
ASSERT (ret == 0);
ASSERT (wc == (wchar_t) 0xBADFACE);
ASSERT (mbsinit (&state));
}
if (argc > 1)
switch (argv[1][0])
{
case '1':
/* Locale encoding is ISO-8859-1 or ISO-8859-15. */
{
char input[] = "B\374\337er"; /* "Büßer" */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'B');
ASSERT (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 1, &state);
ASSERT (ret == 1);
ASSERT (wctob (wc) == (unsigned char) '\374');
ASSERT (mbsinit (&state));
input[1] = '\0';
/* Test support of NULL first argument. */
ret = mbrtowc (NULL, input + 2, 3, &state);
ASSERT (ret == 1);
ASSERT (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 2, 3, &state);
ASSERT (ret == 1);
ASSERT (wctob (wc) == (unsigned char) '\337');
ASSERT (mbsinit (&state));
input[2] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 3, 2, &state);
ASSERT (ret == 1);
ASSERT (wc == 'e');
ASSERT (mbsinit (&state));
input[3] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 4, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'r');
ASSERT (mbsinit (&state));
}
return 0;
case '2':
/* Locale encoding is UTF-8. */
{
char input[] = "B\303\274\303\237er"; /* "Büßer" */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'B');
ASSERT (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 1, &state);
ASSERT (ret == (size_t)(-2));
ASSERT (wc == (wchar_t) 0xBADFACE);
ASSERT (!mbsinit (&state));
input[1] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 2, 5, &state);
ASSERT (ret == 1);
ASSERT (wctob (wc) == EOF);
ASSERT (mbsinit (&state));
input[2] = '\0';
/* Test support of NULL first argument. */
ret = mbrtowc (NULL, input + 3, 4, &state);
ASSERT (ret == 2);
ASSERT (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 3, 4, &state);
ASSERT (ret == 2);
ASSERT (wctob (wc) == EOF);
ASSERT (mbsinit (&state));
input[3] = '\0';
input[4] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 5, 2, &state);
ASSERT (ret == 1);
ASSERT (wc == 'e');
ASSERT (mbsinit (&state));
input[5] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 6, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'r');
ASSERT (mbsinit (&state));
}
return 0;
case '3':
/* Locale encoding is EUC-JP. */
{
char input[] = "<\306\374\313\334\270\354>"; /* "<日本語>" */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == '<');
ASSERT (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 2, &state);
ASSERT (ret == 2);
ASSERT (wctob (wc) == EOF);
ASSERT (mbsinit (&state));
input[1] = '\0';
input[2] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 3, 1, &state);
ASSERT (ret == (size_t)(-2));
ASSERT (wc == (wchar_t) 0xBADFACE);
ASSERT (!mbsinit (&state));
input[3] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 4, 4, &state);
ASSERT (ret == 1);
ASSERT (wctob (wc) == EOF);
ASSERT (mbsinit (&state));
input[4] = '\0';
/* Test support of NULL first argument. */
ret = mbrtowc (NULL, input + 5, 3, &state);
ASSERT (ret == 2);
ASSERT (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 5, 3, &state);
ASSERT (ret == 2);
ASSERT (wctob (wc) == EOF);
ASSERT (mbsinit (&state));
input[5] = '\0';
input[6] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 7, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == '>');
ASSERT (mbsinit (&state));
}
return 0;
case '4':
/* Locale encoding is GB18030. */
{
char input[] = "B\250\271\201\060\211\070er"; /* "Büßer" */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'B');
ASSERT (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 1, &state);
ASSERT (ret == (size_t)(-2));
ASSERT (wc == (wchar_t) 0xBADFACE);
ASSERT (!mbsinit (&state));
input[1] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 2, 7, &state);
ASSERT (ret == 1);
ASSERT (wctob (wc) == EOF);
ASSERT (mbsinit (&state));
input[2] = '\0';
/* Test support of NULL first argument. */
ret = mbrtowc (NULL, input + 3, 6, &state);
ASSERT (ret == 4);
ASSERT (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 3, 6, &state);
ASSERT (ret == 4);
ASSERT (wctob (wc) == EOF);
ASSERT (mbsinit (&state));
input[3] = '\0';
input[4] = '\0';
input[5] = '\0';
input[6] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 7, 2, &state);
ASSERT (ret == 1);
ASSERT (wc == 'e');
ASSERT (mbsinit (&state));
input[5] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 8, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'r');
ASSERT (mbsinit (&state));
}
return 0;
}
return 1;
}
/* Converts a string to an EWF header2
* Returns 1 if successful, 0 if header2 was not set, or -1 on error
*/
int8_t libewf_string_copy_to_header2( LIBEWF_CHAR *string, size_t size_string, EWF_HEADER2 *header2, size_t size_header2 )
{
#ifdef HAVE_WIDE_CHARACTER_TYPE
mbstate_t conversion_state;
#endif
if( string == NULL )
{
LIBEWF_WARNING_PRINT( "libewf_string_copy_to_header2: invalid string.\n" );
return( -1 );
}
if( header2 == NULL )
{
LIBEWF_WARNING_PRINT( "libewf_string_copy_to_header2: invalid header2.\n" );
return( -1 );
}
if( ( size_string > (size_t) SSIZE_MAX ) || ( size_header2 > (size_t) SSIZE_MAX ) )
{
LIBEWF_WARNING_PRINT( "libewf_string_copy_to_header2: invalid size value exceeds maximum.\n" );
return( -1 );
}
if( size_header2 < ( ( size_string * 2 ) + 2 ) )
{
LIBEWF_WARNING_PRINT( "libewf_string_copy_to_header2: header2 too small.\n" );
return( -1 );
}
#ifdef HAVE_WIDE_CHARACTER_TYPE
if( libewf_common_memset( &conversion_state, 0, sizeof( mbstate_t ) ) == NULL )
{
LIBEWF_WARNING_PRINT( "libewf_string_copy_to_header2: unable to clear converion state.\n" );
return( -1 );
}
if( mbsinit( &conversion_state ) == 0 )
{
LIBEWF_WARNING_PRINT( "libewf_string_copy_to_header2: unable to initialize converion state.\n" );
return( -1 );
}
if( wcsrtombs( (char *) &header2[ 2 ], (const wchar_t **) &string, size_header2, &conversion_state ) != ( size_string - 1 ) )
#else
if( libewf_string_copy_ascii_to_utf16( string, size_string, (LIBEWF_CHAR *) header2, size_header2, LIBEWF_STRING_LITTLE_ENDIAN ) != 1 )
#endif
{
LIBEWF_WARNING_PRINT( "libewf_string_copy_to_header2: unable to copy string to header2.\n" );
return( -1 );
}
#ifdef HAVE_WIDE_CHARACTER_TYPE
if( header2[ 4 ] == (EWF_HEADER2) '\0' )
{
header2[ 0 ] = (EWF_HEADER2) 0xfe;
header2[ 1 ] = (EWF_HEADER2) 0xff;
}
else
{
header2[ 0 ] = (EWF_HEADER2) 0xff;
header2[ 1 ] = (EWF_HEADER2) 0xfe;
}
#endif
return( 1 );
}
int
main(int argc, char *argv[])
{
char srcbuf[128];
wchar_t dstbuf[128];
char *src;
mbstate_t s;
/*
* C/POSIX locale.
*/
printf("1..1\n");
/* Simple null terminated string. */
memset(srcbuf, 0xcc, sizeof(srcbuf));
strcpy(srcbuf, "hello");
wmemset(dstbuf, 0xcccc, sizeof(dstbuf) / sizeof(*dstbuf));
src = srcbuf;
memset(&s, 0, sizeof(s));
assert(mbsnrtowcs(dstbuf, (const char **)&src, 6, sizeof(dstbuf) /
sizeof(*dstbuf), &s) == 5);
assert(wcscmp(dstbuf, L"hello") == 0);
assert(dstbuf[6] == 0xcccc);
assert(src == NULL);
/* Simple null terminated string, stopping early. */
memset(srcbuf, 0xcc, sizeof(srcbuf));
strcpy(srcbuf, "hello");
wmemset(dstbuf, 0xcccc, sizeof(dstbuf) / sizeof(*dstbuf));
src = srcbuf;
memset(&s, 0, sizeof(s));
assert(mbsnrtowcs(dstbuf, (const char **)&src, 4, sizeof(dstbuf) /
sizeof(*dstbuf), &s) == 4);
assert(wmemcmp(dstbuf, L"hell", 4) == 0);
assert(dstbuf[5] == 0xcccc);
assert(src == srcbuf + 4);
/* Not enough space in destination buffer. */
memset(srcbuf, 0xcc, sizeof(srcbuf));
strcpy(srcbuf, "hello");
wmemset(dstbuf, 0xcccc, sizeof(dstbuf) / sizeof(*dstbuf));
src = srcbuf;
memset(&s, 0, sizeof(s));
assert(mbsnrtowcs(dstbuf, (const char **)&src, 6, 4, &s) == 4);
assert(wmemcmp(dstbuf, L"hell", 4) == 0);
assert(dstbuf[5] == 0xcccc);
assert(src == srcbuf + 4);
/* Null terminated string, internal dest. buffer */
memset(srcbuf, 0xcc, sizeof(srcbuf));
strcpy(srcbuf, "hello");
src = srcbuf;
memset(&s, 0, sizeof(s));
assert(mbsnrtowcs(NULL, (const char **)&src, 6, 0, &s) == 5);
/* Null terminated string, internal dest. buffer, stopping early */
memset(srcbuf, 0xcc, sizeof(srcbuf));
strcpy(srcbuf, "hello");
src = srcbuf;
memset(&s, 0, sizeof(s));
assert(mbsnrtowcs(NULL, (const char **)&src, 4, 0, &s) == 4);
/* Null terminated string, internal state. */
memset(srcbuf, 0xcc, sizeof(srcbuf));
strcpy(srcbuf, "hello");
wmemset(dstbuf, 0xcccc, sizeof(dstbuf) / sizeof(*dstbuf));
src = srcbuf;
assert(mbsnrtowcs(dstbuf, (const char **)&src, 6, sizeof(dstbuf) /
sizeof(*dstbuf), NULL) == 5);
assert(wcscmp(dstbuf, L"hello") == 0);
assert(dstbuf[6] == 0xcccc);
assert(src == NULL);
/* Null terminated string, internal state, internal dest. buffer. */
memset(srcbuf, 0xcc, sizeof(srcbuf));
strcpy(srcbuf, "hello");
src = srcbuf;
assert(mbsnrtowcs(NULL, (const char **)&src, 6, 0, NULL) == 5);
/* Empty source buffer. */
memset(srcbuf, 0xcc, sizeof(srcbuf));
srcbuf[0] = '\0';
src = srcbuf;
memset(&s, 0, sizeof(s));
wmemset(dstbuf, 0xcccc, sizeof(dstbuf) / sizeof(*dstbuf));
assert(mbsnrtowcs(dstbuf, (const char **)&src, 1, 1, &s) == 0);
assert(dstbuf[0] == 0);
assert(dstbuf[1] == 0xcccc);
assert(src == NULL);
/* Zero length destination buffer. */
memset(srcbuf, 0xcc, sizeof(srcbuf));
strcpy(srcbuf, "hello");
src = srcbuf;
memset(&s, 0, sizeof(s));
wmemset(dstbuf, 0xcccc, sizeof(dstbuf) / sizeof(*dstbuf));
assert(mbsnrtowcs(dstbuf, (const char **)&src, 1, 0, &s) == 0);
assert(dstbuf[0] == 0xcccc);
assert(src == srcbuf);
/* Zero length source buffer. */
memset(srcbuf, 0xcc, sizeof(srcbuf));
src = srcbuf;
memset(&s, 0, sizeof(s));
wmemset(dstbuf, 0xcccc, sizeof(dstbuf) / sizeof(*dstbuf));
assert(mbsnrtowcs(dstbuf, (const char **)&src, 0, 1, &s) == 0);
assert(dstbuf[0] == 0xcccc);
assert(src == srcbuf);
/*
* Japanese (EUC) locale.
*/
assert(strcmp(setlocale(LC_CTYPE, "ja_JP.eucJP"), "ja_JP.eucJP") == 0);
assert(MB_CUR_MAX > 1);
memset(srcbuf, 0xcc, sizeof(srcbuf));
strcpy(srcbuf, "\xA3\xC1 B \xA3\xC3");
src = srcbuf;
memset(&s, 0, sizeof(s));
wmemset(dstbuf, 0xcccc, sizeof(dstbuf) / sizeof(*dstbuf));
assert(mbsnrtowcs(dstbuf, (const char **)&src, 8, sizeof(dstbuf) /
sizeof(*dstbuf), &s) == 5);
assert(dstbuf[0] == 0xA3C1 && dstbuf[1] == 0x20 && dstbuf[2] == 0x42 &&
dstbuf[3] == 0x20 && dstbuf[4] == 0xA3C3 && dstbuf[5] == 0);
assert(src == NULL);
/* Partial character. */
memset(srcbuf, 0xcc, sizeof(srcbuf));
strcpy(srcbuf, "\xA3\xC1 B \xA3\xC3");
src = srcbuf;
memset(&s, 0, sizeof(s));
wmemset(dstbuf, 0xcccc, sizeof(dstbuf) / sizeof(*dstbuf));
assert(mbsnrtowcs(dstbuf, (const char **)&src, 6, sizeof(dstbuf) /
sizeof(*dstbuf), &s) == 4);
assert(src == srcbuf + 6);
assert(!mbsinit(&s));
assert(mbsnrtowcs(dstbuf, (const char **)&src, 1, sizeof(dstbuf) /
sizeof(*dstbuf), &s) == 1);
assert(src == srcbuf + 7);
assert(mbsnrtowcs(dstbuf, (const char **)&src, 1, sizeof(dstbuf) /
sizeof(*dstbuf), &s) == 0);
assert(src == NULL);
printf("ok 1 - mbsnrtowcs()\n");
return (0);
}
int
main (int argc, char *argv[])
{
mbstate_t state;
wchar_t wc;
size_t ret;
/* configure should already have checked that the locale is supported. */
if (setlocale (LC_ALL, "") == NULL)
return 1;
/* Test NUL byte input. */
{
const char *src;
memset (&state, '\0', sizeof (mbstate_t));
src = "";
ret = mbsrtowcs (NULL, &src, 0, &state);
ASSERT (ret == 0);
ASSERT (mbsinit (&state));
src = "";
ret = mbsrtowcs (NULL, &src, 1, &state);
ASSERT (ret == 0);
ASSERT (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
src = "";
ret = mbsrtowcs (&wc, &src, 0, &state);
ASSERT (ret == 0);
ASSERT (wc == (wchar_t) 0xBADFACE);
ASSERT (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
src = "";
ret = mbsrtowcs (&wc, &src, 1, &state);
ASSERT (ret == 0);
ASSERT (wc == 0);
ASSERT (mbsinit (&state));
}
if (argc > 1)
{
int unlimited;
for (unlimited = 0; unlimited < 2; unlimited++)
{
#define BUFSIZE 10
wchar_t buf[BUFSIZE];
const char *src;
mbstate_t temp_state;
{
size_t i;
for (i = 0; i < BUFSIZE; i++)
buf[i] = (wchar_t) 0xBADFACE;
}
switch (argv[1][0])
{
case '1':
/* Locale encoding is ISO-8859-1 or ISO-8859-15. */
{
char input[] = "B\374\337er"; /* "Büßer" */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'B');
ASSERT (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 1, &state);
ASSERT (ret == 1);
ASSERT (wctob (wc) == (unsigned char) '\374');
ASSERT (mbsinit (&state));
input[1] = '\0';
src = input + 2;
temp_state = state;
ret = mbsrtowcs (NULL, &src, unlimited ? BUFSIZE : 1, &temp_state);
ASSERT (ret == 3);
ASSERT (src == input + 2);
ASSERT (mbsinit (&state));
src = input + 2;
ret = mbsrtowcs (buf, &src, unlimited ? BUFSIZE : 1, &state);
ASSERT (ret == (unlimited ? 3 : 1));
ASSERT (src == (unlimited ? NULL : input + 3));
ASSERT (wctob (buf[0]) == (unsigned char) '\337');
if (unlimited)
{
ASSERT (buf[1] == 'e');
ASSERT (buf[2] == 'r');
ASSERT (buf[3] == 0);
ASSERT (buf[4] == (wchar_t) 0xBADFACE);
}
else
ASSERT (buf[1] == (wchar_t) 0xBADFACE);
ASSERT (mbsinit (&state));
}
break;
case '2':
/* Locale encoding is UTF-8. */
{
char input[] = "B\303\274\303\237er"; /* "Büßer" */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'B');
ASSERT (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 1, &state);
ASSERT (ret == (size_t)(-2));
ASSERT (wc == (wchar_t) 0xBADFACE);
ASSERT (!mbsinit (&state));
input[1] = '\0';
src = input + 2;
temp_state = state;
ret = mbsrtowcs (NULL, &src, unlimited ? BUFSIZE : 2, &temp_state);
ASSERT (ret == 4);
ASSERT (src == input + 2);
ASSERT (!mbsinit (&state));
src = input + 2;
ret = mbsrtowcs (buf, &src, unlimited ? BUFSIZE : 2, &state);
ASSERT (ret == (unlimited ? 4 : 2));
ASSERT (src == (unlimited ? NULL : input + 5));
ASSERT (wctob (buf[0]) == EOF);
ASSERT (wctob (buf[1]) == EOF);
if (unlimited)
{
ASSERT (buf[2] == 'e');
ASSERT (buf[3] == 'r');
ASSERT (buf[4] == 0);
ASSERT (buf[5] == (wchar_t) 0xBADFACE);
}
else
ASSERT (buf[2] == (wchar_t) 0xBADFACE);
ASSERT (mbsinit (&state));
}
break;
case '3':
/* Locale encoding is EUC-JP. */
{
char input[] = "<\306\374\313\334\270\354>"; /* "<日本語>" */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == '<');
ASSERT (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 2, &state);
ASSERT (ret == 2);
ASSERT (wctob (wc) == EOF);
ASSERT (mbsinit (&state));
input[1] = '\0';
input[2] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 3, 1, &state);
ASSERT (ret == (size_t)(-2));
ASSERT (wc == (wchar_t) 0xBADFACE);
ASSERT (!mbsinit (&state));
input[3] = '\0';
src = input + 4;
temp_state = state;
ret = mbsrtowcs (NULL, &src, unlimited ? BUFSIZE : 2, &temp_state);
ASSERT (ret == 3);
ASSERT (src == input + 4);
ASSERT (!mbsinit (&state));
src = input + 4;
ret = mbsrtowcs (buf, &src, unlimited ? BUFSIZE : 2, &state);
ASSERT (ret == (unlimited ? 3 : 2));
ASSERT (src == (unlimited ? NULL : input + 7));
ASSERT (wctob (buf[0]) == EOF);
ASSERT (wctob (buf[1]) == EOF);
if (unlimited)
{
ASSERT (buf[2] == '>');
ASSERT (buf[3] == 0);
ASSERT (buf[4] == (wchar_t) 0xBADFACE);
}
else
ASSERT (buf[2] == (wchar_t) 0xBADFACE);
ASSERT (mbsinit (&state));
}
break;
case '4':
/* Locale encoding is GB18030. */
{
char input[] = "B\250\271\201\060\211\070er"; /* "Büßer" */
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
ASSERT (ret == 1);
ASSERT (wc == 'B');
ASSERT (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 1, &state);
ASSERT (ret == (size_t)(-2));
ASSERT (wc == (wchar_t) 0xBADFACE);
ASSERT (!mbsinit (&state));
input[1] = '\0';
src = input + 2;
temp_state = state;
ret = mbsrtowcs (NULL, &src, unlimited ? BUFSIZE : 2, &temp_state);
ASSERT (ret == 4);
ASSERT (src == input + 2);
ASSERT (!mbsinit (&state));
src = input + 2;
ret = mbsrtowcs (buf, &src, unlimited ? BUFSIZE : 2, &state);
ASSERT (ret == (unlimited ? 4 : 2));
ASSERT (src == (unlimited ? NULL : input + 7));
ASSERT (wctob (buf[0]) == EOF);
ASSERT (wctob (buf[1]) == EOF);
if (unlimited)
{
ASSERT (buf[2] == 'e');
ASSERT (buf[3] == 'r');
ASSERT (buf[4] == 0);
ASSERT (buf[5] == (wchar_t) 0xBADFACE);
}
else
ASSERT (buf[2] == (wchar_t) 0xBADFACE);
ASSERT (mbsinit (&state));
}
break;
default:
return 1;
}
}
return 0;
}
return 1;
}
size_t mbrtowc(wchar_t* pwc, const char* s, size_t n, mbstate_t* ps) {
static mbstate_t __private_state;
mbstate_t* state = (ps == NULL) ? &__private_state : ps;
// We should never get to a state which has all 4 bytes of the sequence set.
// Full state verification is done when decoding the sequence (after we have
// all the bytes).
if (mbstate_get_byte(state, 3) != 0) {
return reset_and_return_illegal(EINVAL, state);
}
if (s == NULL) {
s = "";
n = 1;
pwc = NULL;
}
if (n == 0) {
return 0;
}
uint8_t ch;
if (mbsinit(state) && (((ch = static_cast<uint8_t>(*s)) & ~0x7f) == 0)) {
// Fast path for plain ASCII characters.
if (pwc != NULL) {
*pwc = ch;
}
return (ch != '\0' ? 1 : 0);
}
// Determine the number of octets that make up this character
// from the first octet, and a mask that extracts the
// interesting bits of the first octet. We already know
// the character is at least two bytes long.
size_t length;
int mask;
// We also specify a lower bound for the character code to
// detect redundant, non-"shortest form" encodings. For
// example, the sequence C0 80 is _not_ a legal representation
// of the null character. This enforces a 1-to-1 mapping
// between character codes and their multibyte representations.
wchar_t lower_bound;
// The first byte in the state (if any) tells the length.
size_t bytes_so_far = mbstate_bytes_so_far(state);
ch = bytes_so_far > 0 ? mbstate_get_byte(state, 0) : static_cast<uint8_t>(*s);
if ((ch & 0x80) == 0) {
mask = 0x7f;
length = 1;
lower_bound = 0;
} else if ((ch & 0xe0) == 0xc0) {
mask = 0x1f;
length = 2;
lower_bound = 0x80;
} else if ((ch & 0xf0) == 0xe0) {
mask = 0x0f;
length = 3;
lower_bound = 0x800;
} else if ((ch & 0xf8) == 0xf0) {
mask = 0x07;
length = 4;
lower_bound = 0x10000;
} else {
// Malformed input; input is not UTF-8. See RFC 3629.
return reset_and_return_illegal(EILSEQ, state);
}
// Fill in the state.
size_t bytes_wanted = length - bytes_so_far;
size_t i;
for (i = 0; i < MIN(bytes_wanted, n); i++) {
if (!mbsinit(state) && ((*s & 0xc0) != 0x80)) {
// Malformed input; bad characters in the middle of a character.
return reset_and_return_illegal(EILSEQ, state);
}
mbstate_set_byte(state, bytes_so_far + i, *s++);
}
if (i < bytes_wanted) {
return ERR_INCOMPLETE_SEQUENCE;
}
// Decode the octet sequence representing the character in chunks
// of 6 bits, most significant first.
wchar_t wch = mbstate_get_byte(state, 0) & mask;
for (i = 1; i < length; i++) {
wch <<= 6;
wch |= mbstate_get_byte(state, i) & 0x3f;
}
if (wch < lower_bound) {
// Malformed input; redundant encoding.
return reset_and_return_illegal(EILSEQ, state);
}
if ((wch >= 0xd800 && wch <= 0xdfff) || wch == 0xfffe || wch == 0xffff) {
// Malformed input; invalid code points.
return reset_and_return_illegal(EILSEQ, state);
}
if (pwc != NULL) {
*pwc = wch;
}
return reset_and_return(wch == L'\0' ? 0 : bytes_wanted, state);
}
size_t
rpl_mbrtowc (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps)
{
# if MBRTOWC_NULL_ARG2_BUG || MBRTOWC_RETVAL_BUG
if (s == NULL)
{
pwc = NULL;
s = "";
n = 1;
}
# endif
# if MBRTOWC_RETVAL_BUG
{
static mbstate_t internal_state;
/* Override mbrtowc's internal state. We cannot call mbsinit() on the
hidden internal state, but we can call it on our variable. */
if (ps == NULL)
ps = &internal_state;
if (!mbsinit (ps))
{
/* Parse the rest of the multibyte character byte for byte. */
size_t count = 0;
for (; n > 0; s++, n--)
{
wchar_t wc;
size_t ret = mbrtowc (&wc, s, 1, ps);
if (ret == (size_t)(-1))
return (size_t)(-1);
count++;
if (ret != (size_t)(-2))
{
/* The multibyte character has been completed. */
if (pwc != NULL)
*pwc = wc;
return (wc == 0 ? 0 : count);
}
}
return (size_t)(-2);
}
}
# endif
# if MBRTOWC_NUL_RETVAL_BUG
{
wchar_t wc;
size_t ret = mbrtowc (&wc, s, n, ps);
if (ret != (size_t)(-1) && ret != (size_t)(-2))
{
if (pwc != NULL)
*pwc = wc;
if (wc == 0)
ret = 0;
}
return ret;
}
# else
{
# if MBRTOWC_NULL_ARG1_BUG
wchar_t dummy;
if (pwc == NULL)
pwc = &dummy;
# endif
return mbrtowc (pwc, s, n, ps);
}
# endif
}
/**
* ntfs_mbstoucs - convert a multibyte string to a little endian Unicode string
* @ins: input multibyte string buffer
* @outs: on return contains the (allocated) output Unicode string
* @outs_len: length of output buffer in Unicode characters
*
* Convert the input multibyte string @ins, from the current locale into the
* corresponding little endian, 2-byte Unicode string.
*
* If *@outs is NULL, the function allocates the string and the caller is
* responsible for calling free(*@outs); when finished with it.
*
* On success the function returns the number of Unicode characters written to
* the output string *@outs (>= 0), not counting the terminating Unicode NULL
* character. If the output string buffer was allocated, *@outs is set to it.
*
* On error, -1 is returned, and errno is set to the error code. The following
* error codes can be expected:
* EINVAL Invalid arguments (e.g. @ins or @outs is NULL).
* EILSEQ The input string cannot be represented as a Unicode
* string according to the current locale.
* ENAMETOOLONG Destination buffer is too small for input string.
* ENOMEM Not enough memory to allocate destination buffer.
*/
int ntfs_mbstoucs(const char *ins, ntfschar **outs, int outs_len)
{
ntfschar *ucs;
const char *s;
wchar_t wc;
int i, o, cnt, ins_len, ucs_len, ins_size;
#ifdef HAVE_MBSINIT
mbstate_t mbstate;
#endif
if (!ins || !outs) {
errno = EINVAL;
return -1;
}
ucs = *outs;
ucs_len = outs_len;
if (ucs && !ucs_len) {
errno = ENAMETOOLONG;
return -1;
}
/* Determine the size of the multi-byte string in bytes. */
ins_size = strlen(ins);
/* Determine the length of the multi-byte string. */
s = ins;
#if defined(HAVE_MBSINIT)
memset(&mbstate, 0, sizeof(mbstate));
ins_len = mbsrtowcs(NULL, (const char **)&s, 0, &mbstate);
#ifdef __CYGWIN32__
if (!ins_len && *ins) {
/* Older Cygwin had broken mbsrtowcs() implementation. */
ins_len = strlen(ins);
}
#endif
#elif !defined(DJGPP)
ins_len = mbstowcs(NULL, s, 0);
#else
/* Eeek!!! DJGPP has broken mbstowcs() implementation!!! */
ins_len = strlen(ins);
#endif
if (ins_len == -1)
return ins_len;
#ifdef HAVE_MBSINIT
if ((s != ins) || !mbsinit(&mbstate)) {
#else
if (s != ins) {
#endif
errno = EILSEQ;
return -1;
}
/* Add the NULL terminator. */
ins_len++;
if (!ucs) {
ucs_len = ins_len;
ucs = (ntfschar*)malloc(ucs_len * sizeof(ntfschar));
if (!ucs)
return -1;
}
#ifdef HAVE_MBSINIT
memset(&mbstate, 0, sizeof(mbstate));
#else
mbtowc(NULL, NULL, 0);
#endif
for (i = o = cnt = 0; i < ins_size; i += cnt, o++) {
/* Reallocate memory if necessary or abort. */
if (o >= ucs_len) {
ntfschar *tc;
if (ucs == *outs) {
errno = ENAMETOOLONG;
return -1;
}
/*
* We will never get here but hey, it's only a bit of
* extra code...
*/
ucs_len = (ucs_len * sizeof(ntfschar) + 64) & ~63;
tc = (ntfschar*)realloc(ucs, ucs_len);
if (!tc)
goto err_out;
ucs = tc;
ucs_len /= sizeof(ntfschar);
}
/* Convert the multibyte character to a wide character. */
#ifdef HAVE_MBSINIT
cnt = mbrtowc(&wc, ins + i, ins_size - i, &mbstate);
#else
cnt = mbtowc(&wc, ins + i, ins_size - i);
#endif
if (!cnt)
break;
if (cnt == -1)
goto err_out;
if (cnt < -1) {
ntfs_log_trace("Eeek. cnt = %i\n", cnt);
errno = EINVAL;
goto err_out;
}
/* Make sure we are not overflowing the NTFS Unicode set. */
if ((unsigned long)wc >= (unsigned long)(1 <<
(8 * sizeof(ntfschar)))) {
errno = EILSEQ;
goto err_out;
}
/* Convert the CPU wide character to a LE Unicode character. */
ucs[o] = cpu_to_le16(wc);
}
#ifdef HAVE_MBSINIT
/* Make sure we are back in the initial state. */
if (!mbsinit(&mbstate)) {
ntfs_log_trace("Eeek. mbstate not in initial state!\n");
errno = EILSEQ;
goto err_out;
}
#endif
/* Now write the NULL character. */
ucs[o] = cpu_to_le16(L'\0');
if (*outs != ucs)
*outs = ucs;
return o;
err_out:
if (ucs != *outs) {
int eo = errno;
free(ucs);
errno = eo;
}
return -1;
}
/**
* ntfs_upcase_table_build - build the default upcase table for NTFS
* @uc: destination buffer where to store the built table
* @uc_len: size of destination buffer in bytes
*
* ntfs_upcase_table_build() builds the default upcase table for NTFS and
* stores it in the caller supplied buffer @uc of size @uc_len.
*
* Note, @uc_len must be at least 128kiB in size or bad things will happen!
*/
void ntfs_upcase_table_build(ntfschar *uc, u32 uc_len)
{
static int uc_run_table[][3] = { /* Start, End, Add */
{0x0061, 0x007B, -32}, {0x0451, 0x045D, -80}, {0x1F70, 0x1F72, 74},
{0x00E0, 0x00F7, -32}, {0x045E, 0x0460, -80}, {0x1F72, 0x1F76, 86},
{0x00F8, 0x00FF, -32}, {0x0561, 0x0587, -48}, {0x1F76, 0x1F78, 100},
{0x0256, 0x0258, -205}, {0x1F00, 0x1F08, 8}, {0x1F78, 0x1F7A, 128},
{0x028A, 0x028C, -217}, {0x1F10, 0x1F16, 8}, {0x1F7A, 0x1F7C, 112},
{0x03AC, 0x03AD, -38}, {0x1F20, 0x1F28, 8}, {0x1F7C, 0x1F7E, 126},
{0x03AD, 0x03B0, -37}, {0x1F30, 0x1F38, 8}, {0x1FB0, 0x1FB2, 8},
{0x03B1, 0x03C2, -32}, {0x1F40, 0x1F46, 8}, {0x1FD0, 0x1FD2, 8},
{0x03C2, 0x03C3, -31}, {0x1F51, 0x1F52, 8}, {0x1FE0, 0x1FE2, 8},
{0x03C3, 0x03CC, -32}, {0x1F53, 0x1F54, 8}, {0x1FE5, 0x1FE6, 7},
{0x03CC, 0x03CD, -64}, {0x1F55, 0x1F56, 8}, {0x2170, 0x2180, -16},
{0x03CD, 0x03CF, -63}, {0x1F57, 0x1F58, 8}, {0x24D0, 0x24EA, -26},
{0x0430, 0x0450, -32}, {0x1F60, 0x1F68, 8}, {0xFF41, 0xFF5B, -32},
{0}
};
static int uc_dup_table[][2] = { /* Start, End */
{0x0100, 0x012F}, {0x01A0, 0x01A6}, {0x03E2, 0x03EF}, {0x04CB, 0x04CC},
{0x0132, 0x0137}, {0x01B3, 0x01B7}, {0x0460, 0x0481}, {0x04D0, 0x04EB},
{0x0139, 0x0149}, {0x01CD, 0x01DD}, {0x0490, 0x04BF}, {0x04EE, 0x04F5},
{0x014A, 0x0178}, {0x01DE, 0x01EF}, {0x04BF, 0x04BF}, {0x04F8, 0x04F9},
{0x0179, 0x017E}, {0x01F4, 0x01F5}, {0x04C1, 0x04C4}, {0x1E00, 0x1E95},
{0x018B, 0x018B}, {0x01FA, 0x0218}, {0x04C7, 0x04C8}, {0x1EA0, 0x1EF9},
{0}
};
static int uc_byte_table[][2] = { /* Offset, Value */
{0x00FF, 0x0178}, {0x01AD, 0x01AC}, {0x01F3, 0x01F1}, {0x0269, 0x0196},
{0x0183, 0x0182}, {0x01B0, 0x01AF}, {0x0253, 0x0181}, {0x026F, 0x019C},
{0x0185, 0x0184}, {0x01B9, 0x01B8}, {0x0254, 0x0186}, {0x0272, 0x019D},
{0x0188, 0x0187}, {0x01BD, 0x01BC}, {0x0259, 0x018F}, {0x0275, 0x019F},
{0x018C, 0x018B}, {0x01C6, 0x01C4}, {0x025B, 0x0190}, {0x0283, 0x01A9},
{0x0192, 0x0191}, {0x01C9, 0x01C7}, {0x0260, 0x0193}, {0x0288, 0x01AE},
{0x0199, 0x0198}, {0x01CC, 0x01CA}, {0x0263, 0x0194}, {0x0292, 0x01B7},
{0x01A8, 0x01A7}, {0x01DD, 0x018E}, {0x0268, 0x0197},
{0}
};
int i, r;
memset((char*)uc, 0, uc_len);
uc_len >>= 1;
if (uc_len > 65536)
uc_len = 65536;
for (i = 0; (u32)i < uc_len; i++)
uc[i] = i;
for (r = 0; uc_run_table[r][0]; r++)
for (i = uc_run_table[r][0]; i < uc_run_table[r][1]; i++)
uc[i] += uc_run_table[r][2];
for (r = 0; uc_dup_table[r][0]; r++)
for (i = uc_dup_table[r][0]; i < uc_dup_table[r][1]; i += 2)
uc[i + 1]--;
for (r = 0; uc_byte_table[r][0]; r++)
uc[uc_byte_table[r][0]] = uc_byte_table[r][1];
}
/**
* ntfs_str2ucs - convert a string to a valid NTFS file name
* @s: input string
* @len: length of output buffer in Unicode characters
*
* Convert the input @s string into the corresponding little endian,
* 2-byte Unicode string. The length of the converted string is less
* or equal to the maximum length allowed by the NTFS format (255).
*
* If @s is NULL then return AT_UNNAMED.
*
* On success the function returns the Unicode string in an allocated
* buffer and the caller is responsible to free it when it's not needed
* anymore.
*
* On error NULL is returned and errno is set to the error code.
*/
ntfschar *ntfs_str2ucs(const char *s, int *len)
{
ntfschar *ucs = NULL;
if (s && ((*len = ntfs_mbstoucs(s, &ucs, 0)) == -1)) {
ntfs_log_perror("Couldn't convert '%s' to Unicode", s);
return NULL;
}
if (*len > 0xff) {
free(ucs);
errno = ENAMETOOLONG;
return NULL;
}
if (!ucs || !*len) {
ucs = AT_UNNAMED;
*len = 0;
}
return ucs;
}
int
fnmatch (const char *pattern, const char *string, int flags)
{
# if HANDLE_MULTIBYTE
# define ALLOCA_LIMIT 2000
if (__builtin_expect (MB_CUR_MAX, 1) != 1)
{
mbstate_t ps;
size_t patsize;
size_t strsize;
size_t totsize;
wchar_t *wpattern;
wchar_t *wstring;
int res;
/* Calculate the size needed to convert the strings to
wide characters. */
memset (&ps, '\0', sizeof (ps));
patsize = mbsrtowcs (NULL, &pattern, 0, &ps) + 1;
if (__builtin_expect (patsize != 0, 1))
{
assert (mbsinit (&ps));
strsize = mbsrtowcs (NULL, &string, 0, &ps) + 1;
if (__builtin_expect (strsize != 0, 1))
{
assert (mbsinit (&ps));
totsize = patsize + strsize;
if (__builtin_expect (! (patsize <= totsize
&& totsize <= SIZE_MAX / sizeof (wchar_t)),
0))
{
errno = ENOMEM;
return -1;
}
/* Allocate room for the wide characters. */
if (__builtin_expect (totsize < ALLOCA_LIMIT, 1))
wpattern = (wchar_t *) alloca (totsize * sizeof (wchar_t));
else
{
wpattern = malloc (totsize * sizeof (wchar_t));
if (__builtin_expect (! wpattern, 0))
{
errno = ENOMEM;
return -1;
}
}
wstring = wpattern + patsize;
/* Convert the strings into wide characters. */
mbsrtowcs (wpattern, &pattern, patsize, &ps);
assert (mbsinit (&ps));
mbsrtowcs (wstring, &string, strsize, &ps);
res = internal_fnwmatch (wpattern, wstring, wstring + strsize - 1,
flags & FNM_PERIOD, flags);
if (__builtin_expect (! (totsize < ALLOCA_LIMIT), 0))
free (wpattern);
return res;
}
}
}
# endif /* HANDLE_MULTIBYTE */
return internal_fnmatch (pattern, string, string + strlen (string),
flags & FNM_PERIOD, flags);
}
int
main (int argc, char *argv[])
{
mbstate_t state;
wchar_t wc;
size_t ret;
int i;
/* configure should already have checked that the locale is supported. */
if (setlocale (LC_ALL, "") == NULL) {
fprintf(stderr, "unable to set standard locale\n");
return 1;
}
/* Test zero-length input. */
printf("zero-length input ...\n");
{
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "x", 0, &state);
/* gnulib's implementation returns (size_t)(-2).
The AIX 5.1 implementation returns (size_t)(-1).
glibc's implementation returns 0. */
assert (ret == (size_t)(-2) || ret == (size_t)(-1) || ret == 0);
assert (mbsinit (&state));
}
/* Test NUL byte input. */
printf("NUL byte input ...\n");
{
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, "", 1, &state);
assert (ret == 0);
assert (wc == 0);
assert (mbsinit (&state));
ret = mbrtowc (NULL, "", 1, &state);
assert (ret == 0);
assert (mbsinit (&state));
}
/* Test single-byte input. */
printf("single-byte input ...\n");
{
char buf[1];
int c;
memset (&state, '\0', sizeof (mbstate_t));
for (c = 0; c < 0x100; c++)
switch (c)
{
case '\t': case '\v': case '\f':
case ' ': case '!': case '"': case '#': case '%':
case '&': case '\'': case '(': case ')': case '*':
case '+': case ',': case '-': case '.': case '/':
case '0': case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
case ':': case ';': case '<': case '=': case '>':
case '?':
case 'A': case 'B': case 'C': case 'D': case 'E':
case 'F': case 'G': case 'H': case 'I': case 'J':
case 'K': case 'L': case 'M': case 'N': case 'O':
case 'P': case 'Q': case 'R': case 'S': case 'T':
case 'U': case 'V': case 'W': case 'X': case 'Y':
case 'Z':
case '[': case '\\': case ']': case '^': case '_':
case 'a': case 'b': case 'c': case 'd': case 'e':
case 'f': case 'g': case 'h': case 'i': case 'j':
case 'k': case 'l': case 'm': case 'n': case 'o':
case 'p': case 'q': case 'r': case 's': case 't':
case 'u': case 'v': case 'w': case 'x': case 'y':
case 'z': case '{': case '|': case '}': case '~':
/* c is in the ISO C "basic character set". */
buf[0] = c;
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, buf, 1, &state);
assert (ret == 1);
assert (wc == c);
assert (mbsinit (&state));
ret = mbrtowc (NULL, buf, 1, &state);
assert (ret == 1);
assert (mbsinit (&state));
break;
}
}
/* Test special calling convention, passing a NULL pointer. */
printf("special calling convention, passing NULL ...\n");
{
memset (&state, '\0', sizeof (mbstate_t));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, NULL, 5, &state);
assert (ret == 0);
assert (wc == (wchar_t) 0xBADFACE);
assert (mbsinit (&state));
}
for (i = '1'; i <= '4'; ++i) {
switch (i)
{
case '1':
/* Locale encoding is ISO-8859-1 or ISO-8859-15. */
printf("ISO8859-1 ...\n");
{
char input[] = "B\374\337er"; /* "Büßer" */
memset (&state, '\0', sizeof (mbstate_t));
if (setlocale (LC_ALL, "en_US.ISO8859-1") == NULL) {
fprintf(stderr, "unable to set ISO8859-1 locale, skipping\n");
break;
}
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
assert (ret == 1);
assert (wc == 'B');
assert (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 1, &state);
assert (ret == 1);
assert (wctob (wc) == (unsigned char) '\374');
assert (mbsinit (&state));
input[1] = '\0';
/* Test support of NULL first argument. */
ret = mbrtowc (NULL, input + 2, 3, &state);
assert (ret == 1);
assert (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 2, 3, &state);
assert (ret == 1);
assert (wctob (wc) == (unsigned char) '\337');
assert (mbsinit (&state));
input[2] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 3, 2, &state);
assert (ret == 1);
assert (wc == 'e');
assert (mbsinit (&state));
input[3] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 4, 1, &state);
assert (ret == 1);
assert (wc == 'r');
assert (mbsinit (&state));
}
break;
case '2':
/* Locale encoding is UTF-8. */
printf("UTF-8 ...\n");
{
char input[] = "B\303\274\303\237er"; /* "Büßer" */
memset (&state, '\0', sizeof (mbstate_t));
if (setlocale (LC_ALL, "en_US.UTF-8") == NULL) {
fprintf(stderr, "unable to set UTF-8 locale, skipping\n");
break;
}
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
assert (ret == 1);
assert (wc == 'B');
assert (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 1, &state);
assert (ret == (size_t)(-2));
assert (wc == (wchar_t) 0xBADFACE);
assert (!mbsinit (&state));
input[1] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 2, 5, &state);
assert (ret == 1);
assert (wctob (wc) == EOF);
assert (mbsinit (&state));
input[2] = '\0';
/* Test support of NULL first argument. */
ret = mbrtowc (NULL, input + 3, 4, &state);
assert (ret == 2);
assert (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 3, 4, &state);
assert (ret == 2);
assert (wctob (wc) == EOF);
assert (mbsinit (&state));
input[3] = '\0';
input[4] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 5, 2, &state);
assert (ret == 1);
assert (wc == 'e');
assert (mbsinit (&state));
input[5] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 6, 1, &state);
assert (ret == 1);
assert (wc == 'r');
assert (mbsinit (&state));
}
break;
case '3':
/* Locale encoding is EUC-JP. */
printf("EUC-JP ...\n");
{
char input[] = "<\306\374\313\334\270\354>"; /* "<日本語>" */
memset (&state, '\0', sizeof (mbstate_t));
if (setlocale (LC_ALL, "en_US.EUC-JP") == NULL) {
fprintf(stderr, "unable to set EUC-JP locale, skipping\n");
break;
}
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
assert (ret == 1);
assert (wc == '<');
assert (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 2, &state);
assert (ret == 2);
assert (wctob (wc) == EOF);
assert (mbsinit (&state));
input[1] = '\0';
input[2] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 3, 1, &state);
assert (ret == (size_t)(-2));
assert (wc == (wchar_t) 0xBADFACE);
assert (!mbsinit (&state));
input[3] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 4, 4, &state);
assert (ret == 1);
assert (wctob (wc) == EOF);
assert (mbsinit (&state));
input[4] = '\0';
/* Test support of NULL first argument. */
ret = mbrtowc (NULL, input + 5, 3, &state);
assert (ret == 2);
assert (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 5, 3, &state);
assert (ret == 2);
assert (wctob (wc) == EOF);
assert (mbsinit (&state));
input[5] = '\0';
input[6] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 7, 1, &state);
assert (ret == 1);
assert (wc == '>');
assert (mbsinit (&state));
}
break;
case '4':
/* Locale encoding is GB18030. */
printf("GB18030 ...\n");
{
char input[] = "B\250\271\201\060\211\070er"; /* "Büßer" */
memset (&state, '\0', sizeof (mbstate_t));
if (setlocale (LC_ALL, "en_US.GB18030") == NULL) {
fprintf(stderr, "unable to set GB18030 locale, skipping\n");
break;
}
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input, 1, &state);
assert (ret == 1);
assert (wc == 'B');
assert (mbsinit (&state));
input[0] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 1, 1, &state);
assert (ret == (size_t)(-2));
assert (wc == (wchar_t) 0xBADFACE);
assert (!mbsinit (&state));
input[1] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 2, 7, &state);
assert (ret == 1);
assert (wctob (wc) == EOF);
assert (mbsinit (&state));
input[2] = '\0';
/* Test support of NULL first argument. */
ret = mbrtowc (NULL, input + 3, 6, &state);
assert (ret == 4);
assert (mbsinit (&state));
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 3, 6, &state);
assert (ret == 4);
assert (wctob (wc) == EOF);
assert (mbsinit (&state));
input[3] = '\0';
input[4] = '\0';
input[5] = '\0';
input[6] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 7, 2, &state);
assert (ret == 1);
assert (wc == 'e');
assert (mbsinit (&state));
input[5] = '\0';
wc = (wchar_t) 0xBADFACE;
ret = mbrtowc (&wc, input + 8, 1, &state);
assert (ret == 1);
assert (wc == 'r');
assert (mbsinit (&state));
}
break;
}
}
return 0;
}